Part I — Situation overview

The Ministry of Finance published its preliminary public finance bulletin on 8 September 2026, according to which the deficit of the central subsector was 2,311.2 billion forints in August — on the available data the highest monthly deficit figure ever measured. The cumulative balance for the first eight months thereby rose to minus 5,169.1 billion forints. In the same communication the ministry also gave the explanation: the August deficit was caused primarily by pre-financing linked to the Recovery and Resilience Facility (RRF — the EU recovery fund), which increased expenditure by 2,186.9 billion forints. Cleaned of this one-off item, the August deficit would have been 124.3 billion forints, the lowest eighth-month deficit figure of recent years. The ministry also indicated that the related EU revenues are expected to arrive in December, and that it will then be worth correcting the monthly balance in a similar way — with the opposite sign. The corrected cumulative deficit is thus 2,982.2 billion forints, which fits the pattern of earlier years.

On the day of the publication, in parliament, the general debate on the amendment of the 2026 budget began with the opening speech of András Kármán, the finance minister. The minister justified the untenability of the originally planned deficit target corresponding to 3.7 per cent of gross domestic product (GDP) by reference to the previous cabinet’s management, uncovered election spending, a miscalculated growth path, overpriced state contracts and EU funds built into the budget but not realistically to be expected. According to the ministry’s screening, without intervention this year’s deficit would have reached 8.3 per cent of GDP; with the measures submitted this is reduced to 7.5 per cent. The amendment contains 403 billion forints of specific expenditure reductions, of which 88 billion forints comes from breaking overpricing. Under the amended plan government debt rises from 74.6 per cent at the end of last year to 77.5 per cent by the end of 2026. At the same time the minister put it unambiguously: a deficit of 7.5 per cent is “not acceptable even in the short run”, in 2027 a significant deficit reduction will follow, and in parallel with the preparation of the 2027 budget the government will also submit its medium-term fiscal-structural plan.

MIAK’s reading: behind the record deficit stand two items that must be sharply separated — the EU pre-financing, which is a timing item and therefore returns, and the structural deficit, which does not. The political debate blurs the two on both sides: the government side points to the pre-financing, Fidesz to the nominal record, and both statements are formally true. The problem is not which figure is larger, but that at present only the ministry’s textual commentary says what each of them means. A deficit figure becomes policy information when its breakdown is public and machine-processable as well — without this, every monthly publication remains the terrain of a contest of commentaries.

Part II — Foundations in the literature

Behind the present data dispute stand three items. The International Monetary Fund (IMF) publication World Economic Outlook 2025 treats the quality of data publication as a public good, and names those failings — incomplete coverage, opaque methodology, infrequent publication — which undermine accountability; the same volume identifies medium-term fiscal frameworks with pre-announced, clear reference points as the condition of a credible debt path. Carmen Reinhart and Kenneth Rogoff, the Harvard economists who worked up eight centuries of the history of financial crises, show in their work This Time Is Different (2009) what happens when part of the budgetary picture is regularly left out of the publicly accessible data. The “forgotten history” of hard-to-access domestic debt was able to mislead markets and institutions precisely because it did not feature in the usual statistics. And the anchoring effect described by Daniel Kahneman, the psychologist awarded the Nobel Memorial Prize in Economics for his research in decision theory, in his volume Thinking, Fast and Slow (2011), explains why it is not neutral whether the news starts with the figure of 2,311 billion or that of 124 billion: the value communicated first influences the estimate even when the reader knows that it is not the relevant one. The detailed treatment of the literature — author by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures.

3.1 A separated, machine-readable monthly time series of one-off and structural items (from the October 2026 publication)

The Ministry of Finance has announced that from October 2026 it will publish the preliminary bulletin on the third working day of the month — instead of the eighth as before — and will attach a table alongside the textual assessment. In MIAK’s view this is the right direction, and it proposes that from that same publication the ministry supplement it with one element: let it break down the monthly balance in a machine-readable file into at least three lines — one-off items of a timing character (for example the EU pre-financing), lasting expenditure and revenue processes, and interest expenditure. Let the breakdown appear not in commentary but in a tabular data series, uploaded retrospectively for at least five years so that the time series is comparable. This is the direct application of programme point G1 — the data-driven budget — and in its technical implementation it follows the logic of the A1 public money dashboard. The relevant finding of the IMF (see 6.4.1) records precisely that incomplete or hard-to-interpret publication in itself impairs the effectiveness of policy — the present August figure illustrates exactly this.

3.2 A spending review timetable alongside the 2027 consolidation promise (by the submission of the 2027 budget)

The finance minister stated that in 2027 a significant deficit reduction will follow, and that a medium-term fiscal-structural plan is being prepared. MIAK proposes that by the submission of the 2027 budget a public timetable should also go with this promise: which chapters’ screening starts when, who carries it out, and by when results are expected. Programme point G21 supplies the methodology for this — a zero-based review of every budget item every three years, with the question of whether, if we were not doing it today, we would start it — while programme point KI8 supplies the frame for measuring administrative efficiency. Without a timetable the consolidation promise remains unverifiable, and experience shows that withdrawals in such cases hit the items most easily deferred — investment and maintenance — not the least useful ones. After the debt peak of 77.5 per cent, the G23 debt sustainability framework would perform precisely that early warning function which signals a turn in the path in good time.

3.3 Separate presentation of the December EU revenue and subsequent verification (by the end of the first quarter of 2027)

According to the ministry’s indication, the revenues linked to the RRF programmes will arrive in December, and the monthly balance will then be worth reading with a similar correction. MIAK proposes that this receipt should not be dissolved into the aggregate December figure: let the ministry publish item by item which programme was reimbursed with what amount and when, and how the amount actually received relates to the 2,186.9 billion forints pre-financed in August. By the end of the first quarter of 2027 let a subsequent verification of this be prepared according to the impact assessment logic of programme point G20. This is not a formality: naming something a one-off item is a credible statement only if it can be demonstrated afterwards that it really did return. Reinhart and Rogoff’s warning (see 6.4.2) refers precisely to the fact that the risk of “this is only timing” explanations does not lie in their being false — but in their remaining unchecked.

The three proposals are bound together by the same principle: a budgetary figure is not in itself information; it becomes so only if its breakdown, its antecedents and its subsequent verification are public as well. The monthly breakdown says what happened. The timetable says what comes next. The verification says whether what the explanation asserted was true. None of them restricts the government in what it spends on — only in making its explanation unverifiable after the event.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy Publishing one-off and lasting items separately reduces market uncertainty around the monthly data; the faster publication on the third working day in itself improves predictability If the December EU revenue slips, the state will have to finance the deficit now generated for a longer period, which raises the interest burden; the debt peak of 77.5 per cent will move to the centre of rating agency attention
Public administration The spending review timetable places chapter managers in a frame that can be planned in advance, turning tightening into a decision on priorities instead of a linear withdrawal Zero-based screening ties up significant capacity at the ministries, and work carried out in parallel with the 2027 planning may impair the quality of both processes
Publicity and trust The machine-readable breakdown ends the monopoly on commentary: independent analysts can work from the same data, which turns the debate into a question of fact Detailed data produces more debate in the short run, not less; this creates a political temptation to narrow the publication or to postpone the breakdown

The main tension of the package runs between honesty and credibility. Stating the “no-intervention” path of 8.3 per cent and the amended deficit target of 7.5 per cent was an honest step, and MIAK regards it as such: a budget built on real numbers is a better starting point than maintaining an untenable plan. At the same time every government that raises a deficit target by reference to the situation inherited from its predecessor uses up a one-off opportunity — the same argument will not be available a second time. The proposal tips towards the risk side if the 2027 consolidation is left without a timetable, because then today’s honesty will retrospectively become the justification for a lasting increase in the deficit target. The second question for consideration is the level of detail of the breakdown: classifying an item as one-off is itself a decision, and every such category offers scope for abuse if it is not fixed what qualifies as one. That is why the classification rule has to be published together with the breakdown, and the December verification has to be carried out — the credibility of the category comes from subsequent checking, not from the label.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK proposes four performance indicators (KPIs, in full: Key Performance Indicators) from which it will be visible in 6, 12 and 24 months whether the renewal of budgetary communication has been substantive:

  • The appearance of a machine-readable breakdown in the monthly bulletin: the suggested target is that from the October 2026 publication one-off, lasting and interest items should feature in separate data series, uploaded retrospectively for at least five years.
  • The correspondence between the December EU revenue and the August pre-financing: it is worth following to what extent the amount actually received covers the 2,186.9 billion forints pre-financed, and how large a share slips to 2027.
  • The path of the primary deficit — excluding interest expenditure: this indicator says whether the correction is one-off or lasting; it is worth following on the basis of the figures of the 2027 budget and the medium-term plan.
  • The publication of the spending review timetable: the suggested deadline is the submission of the 2027 budget; the indicator is binary, but the credibility of the consolidation promise depends on it.

5.2 Summary

MIAK’s request in a single sentence: let the Ministry of Finance make machine-readable the breakdown that it publishes today in textual commentary. Concretely: a separated monthly time series of one-off and lasting items from the October 2026 publication, a spending review timetable by the submission of the 2027 budget, and an itemised presentation of the December EU revenue with subsequent verification by the end of the first quarter of 2027. None of these requests is about the government spending less — they are about the monthly figure being interpretable in itself.

Two MIAK foundational values are in play here. One is data-drivenness: a figure of 2,311 billion and a figure of 124 billion refer to the same month, and today what decides which the reader sees is whose commentary they read. This is not a communication problem but a data-structure problem, and it has a data-structure solution. The other is accountability: MIAK asks of the present government the same breakdown that it asked of the previous one — and the yardstick tightens only in so far as the ministry itself announced a transparency improvement. From those who undertake faster publication, a more detailed breakdown may rightly be expected too.

Part VI — Reasoning and further sources

6.1 The framing of the press by spectrum

The framing of the economic band was purely methodological, and it produced the day’s most precise reading. Portfolio indicated even in the headline that “there is a twist in the story”, and the backbone of the article was the juxtaposition of the raw and the corrected figure: alongside the record of 2,311 billion it immediately placed the EU pre-financing of 2,187 billion and the cleaned value of 124 billion, then published the cumulative figure in both versions as well. The same paper separately highlighted that the deficit stands at 72.7 per cent of the raised annual plan of 7,110 billion forints, which is not exceptional at this stage of the year. This framing shifts the debate from the size of the figure to the structure of the figure — precisely where MIAK’s proposal is also directed.

The liberal-left band approached the same data from the side of shock, but did not suppress the explanation. 444.hu opened with the formulation of a “historically dismal figure”, and recalled that the prime minister had days earlier indicated the exceptional August deficit himself; the article then set out in detail the effect of the EU pre-financing, and the expectation of a December surplus as well. The same paper published a separate report on the parliamentary debate, in which, alongside the speeches of the governing parties and the opposition, the assessment of the president of the Fiscal Council also featured: in the body’s view the amended act complies with the legislation, but carries risks and requires greater caution than usual. Telex placed the Hungarian situation in a world economic frame, through the global rise in government bond yields. 24.hu remained factual: it published the data of the ministry report without commentary.

The conservative band placed the nominal record at the centre and tied responsibility to the present government. Magyar Nemzet published the figure under the headline “The government put together a record deficit in August” — though the article also contained the explanation relating to the EU pre-financing, the uncertainty of the December receipt and the debt path rising from 74.6 to 77.5 per cent. Another piece set out Fidesz’s three-item summary, according to which July’s increase in debt and the highest eight-month deficit since 2010 show the failure of government performance. By MIAK’s yardstick this framing is substantive and must be answered on one point: the December EU receipt is indeed not guaranteed, and if it slips, the state will have to finance the deficit now generated for a longer period. MIAK’s answer to this is that it is precisely for this reason that the itemised December statement and the subsequent verification are needed — raising the risk is legitimate, and handling it is not a rhetorical but a data-publication question. At the same time MIAK does not regard conversions of the “3.5 billion forints of indebtedness an hour” type as a policy statement: a breakdown per unit of time illustrates the order of magnitude, and says nothing about the structure of the deficit.

6.2 Facts and data

Data Value Source
August deficit (central subsector) 2,311.2 billion HUF — the highest monthly figure ever measured Ministry of Finance / Portfolio, 8 September 2026
The effect of the RRF pre-financing in August +2,186.9 billion HUF of expenditure Ministry of Finance / 24.hu, 8 September 2026
The August deficit cleaned of the one-off item 124.3 billion HUF Ministry of Finance / Portfolio, 8 September 2026
Cumulative deficit after 8 months 5,169.1 billion HUF Ministry of Finance / Portfolio, 8 September 2026
Corrected cumulative deficit after 8 months 2,982.2 billion HUF Portfolio, 8 September 2026
The amended annual cash-flow deficit plan 7,110 billion HUF (from the original 4,218 billion HUF) Portfolio, 8 September 2026
The cumulative deficit as a proportion of the amended plan 72.7% Portfolio, 8 September 2026
Original deficit target (as a share of GDP) 3.7% 444.hu, 8 September 2026
The no-intervention deficit path according to the ministry’s screening 8.3% Portfolio, 8 September 2026
Amended deficit target 7.5% Portfolio, 8 September 2026
The expenditure reduction of the amendment 403 billion HUF, of which 88 billion HUF from breaking overpricing Portfolio, 8 September 2026
Expenditure and revenue appropriations after the amendment expenditure 43,781 → 49,025 billion HUF; revenue 39,562 → 41,915 billion HUF 444.hu, 8 September 2026
Government debt as a share of GDP end of 2025: 74.6% → end of 2026 (plan): 77.5% Magyar Nemzet / Portfolio, 8 September 2026
Social package in the amendment 63 billion HUF (school-start support, VAT exemption for prescription medicines) Portfolio / Magyar Nemzet, 8 September 2026
The change in the publication regime from October 2026 on the third working day of the month, with a tabular annex (previously the eighth) Ministry of Finance / 444.hu, 8 September 2026

Two lines require a separate note. The original annual deficit plan appears in two forms in the sources: the cash-flow value of 4,218 billion forints and the deficit target corresponding to 3.7 per cent of GDP under EU methodology follow different accounting approaches, so the two figures are not conversions of each other. And the value of 8.3 per cent is not an actual figure but an estimate deriving from the Ministry of Finance’s screening and relating to a no-intervention path — which is why the table marks its source.

6.3 Policy dimensions

  • Economy (programme points) — the structure of budgetary data publication and the foundations of consolidation: G1 (data-driven budget), G20 (subsequent impact assessment), G21 (systematic review of state expenditure) and G23 (debt sustainability framework) together provide the skeleton of the proposal;
  • Transparency and anti-corruption policy (programme points) — the form of publication: A1 (public money dashboard) provides the technical implementation of the machine-readable monthly breakdown;
  • Public administration and e-government (programme points) — the methodology of the review: KI8 (efficiency measurement in public administration) provides the frame in which the screening of chapters becomes a decision on priorities rather than a linear withdrawal.

6.4 Literature in detail

6.4.1 International Monetary Fund: World Economic Outlook 2025

The chapter of the report containing policy recommendations treats institutional quality and data publication in the same line as the fiscal framework. The most important finding from the point of view of the topic:

“High-quality, timely and professionally independent data are a public good: they reduce uncertainty and improve private sector planning and policy formation. By contrast, weak data governance — gaps in coverage, opaque methodology, infrequent publication or politically influenced revisions — undermines accountability and blunts the effectiveness of policy.”

On the credibility of the debt path the same chapter writes that governments should publish medium-term fiscal frameworks with clear reference points, a pre-announced adjustment path and plans for handling shocks, because together these strengthen market confidence and reduce the risk premium. It also warns separately that fiscal strategies built on a favourable baseline or on exceptional growth are themselves sources of vulnerability.

In the Hungarian situation this underpins two statements. On the one hand, the Ministry of Finance’s present step — faster publication and a tabular annex — points precisely in the direction quoted, which is why MIAK’s proposal asks not for a change of direction but for the extension of the direction already taken. On the other hand, the credibility of the 2027 consolidation promise comes not from the firmness of the announcement but from the pre-announced adjustment path placed alongside it — and that is what is still missing today.

📖 Source: International Monetary Fund: World Economic Outlook 2025

6.4.2 Carmen Reinhart and Kenneth Rogoff: This Time Is Different

One of the central claims of the authors’ research is that in the recurrence of crises the leading role is played not by bad intent but by incomplete data and the self-justification that feeds on it. Writing about the history of domestic debt, they record:

“Because historical data on domestic debt are so difficult to obtain, they have been ignored in empirical studies of the debt and inflation of developing countries.”

From this they derive the mechanism of “this time is different” thinking: market participants and institutions alike drew conclusions that rested only on the available, incomplete statistics — and the error lay not in the reasoning but in the database. The authors add that in such cases the actors typically accept the favourable interpretation, and construct after the fact the difference that would “this time” exempt them from the historical pattern.

The Hungarian case is a milder variant of this pattern, but identical in structure. In the case of the August deficit the “only timing” explanation is probably correct — the ministry’s communication is detailed and verifiable. The risk lies not in this single month, but in the “one-off item” category becoming usable without confirmation. That is why MIAK asks for the itemised December statement: not to refute the explanation, but so that the category should remain credible the next time too.

📖 Source: Carmen Reinhart – Kenneth Rogoff: This Time Is Different

6.4.3 Daniel Kahneman: Thinking, Fast and Slow

The eleventh chapter of Kahneman’s volume is about the anchoring effect — the phenomenon whereby a number encountered before estimating a quantity influences the judgment even when it is obviously irrelevant. In the author’s formulation:

“The phenomenon we were studying is so common and so important in the everyday world that you should know its name: it is an anchoring effect. It occurs when people consider a particular value for an unknown quantity before estimating that quantity.”

Kahneman emphasises that the effect operates even when the starting number cannot reasonably be justified, and that the adjustment — that is, moving away from the anchor — typically stops too early. The phenomenon therefore does not stem from ignorance: the reader may know that the anchor is not the relevant number, and their estimate will still remain close to it.

This explains why the order of publication is not a neutral question. If the news starts with the record of 2,311 billion, and the correction of 2,187 billion comes only afterwards, the reader’s estimate stays closer to the record — and vice versa. This is precisely why MIAK’s proposal does not ask for the regulation of commentaries: the machine-readable breakdown, published simultaneously, is the only solution that works not through the order but through the structure of the data.

📖 Source: Daniel Kahneman: Thinking, Fast and Slow

6.5 International comparison

Pre-financing linked to EU recovery funds is not a Hungarian peculiarity: member states typically have to advance the payments of the programmes themselves, and the Commission’s reimbursement arrives afterwards, once the milestones have been met. This structure produces the same accounting tension in every beneficiary member state: the cash-flow balance spikes periodically, while the indicator under EU methodology does not move in the same way. Those member states that show EU programme items on a separate line in the monthly cash-flow publication from the outset resolve this tension technically — not with an explanation but with a data structure. Hungarian practice today inclines towards explanation, and MIAK’s proposal asks precisely for this shift.

The second point of comparison is the speed of publication. The Ministry of Finance’s announcement that from October the bulletin will arrive on the third working day of the month is an advance in international comparison too: publishing monthly cash-flow data at this pace is characteristic of the better half of EU member states. Speed in itself, however, improves predictability only if the structure is stable as well — a rapidly arriving data series whose breakdown changes from year to year is worth less than a slower but comparable time series. The request for retrospective uploading is therefore not an incidental element but the essence of the proposal.

The third consideration concerns the role of medium-term plans. Since the 2024 reform of the EU fiscal framework, member states submit medium-term fiscal-structural plans that fix a net expenditure path. The Hungarian government’s announcement that it will submit such a plan in parallel with the 2027 budget fits into this order. The value of the plan comes from its being verifiable from year to year: the IMF recommendation quoted also refers to that credibility which derives not from the size of the commitment but from the subsequent verifiability of its fulfilment.

Economy

  • G1 — Data-driven budget
  • G20 — Economic policy impact assessment system
  • G21 — Systematic review of state expenditure
  • G23 — Government debt sustainability framework

Transparency and anti-corruption policy

  • A1 — Public money dashboard

Public administration and e-government

  • KI8 — Drucker-style efficiency measurement in public administration

Suggested new programme point: A classification standard for one-off items — for the Economy area: items classified as one-off in the monthly public finance publication should be subject to a public classification criterion and a compulsory subsequent verification.

6.7 List of sources

Press sources (MIAK press monitor, 9 September 2026 — topic 3):

Knowledge base references (specialist books):

  • 📖 International Monetary Fund: World Economic Outlook 2025
  • 📖 Carmen Reinhart – Kenneth Rogoff: This Time Is Different
  • 📖 Daniel Kahneman: Thinking, Fast and Slow

MIAK internal materials:

  • MIAK policy area: Economy (programme points; programme point ID: G1, G20, G21, G23)
  • MIAK policy area: Transparency and anti-corruption policy (programme points; programme point ID: A1)
  • MIAK policy area: Public administration and e-government (programme points; programme point ID: KI8)
  • MIAK press monitor, 9 September 2026 — topic 3, score: 90/100

Supplementary public data sources:

  • Ministry of Finance — monthly preliminary public finance bulletin
  • Fiscal Council — opinion on the amendment of the 2026 budget
  • Eurostat — member state data reporting linked to the excessive deficit procedure

Generation metadata